Amazon UK - Exploratory Data Analysis¶

In [7]:
import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
import seaborn as sns
import plotly.express as px
In [15]:
df_read = pd.read_csv('Desktop/amz_uk_processed_data.csv')
In [16]:
df_read.head(3)
Out[16]:
asin title imgUrl productURL stars reviews price isBestSeller boughtInLastMonth categoryName
0 B09B96TG33 Echo Dot (5th generation, 2022 release) | Big ... https://m.media-amazon.com/images/I/71C3lbbeLs... https://www.amazon.co.uk/dp/B09B96TG33 4.7 15308 21.99 False 0 Hi-Fi Speakers
1 B01HTH3C8S Anker Soundcore mini, Super-Portable Bluetooth... https://m.media-amazon.com/images/I/61c5rSxwP0... https://www.amazon.co.uk/dp/B01HTH3C8S 4.7 98099 23.99 True 0 Hi-Fi Speakers
2 B09B8YWXDF Echo Dot (5th generation, 2022 release) | Big ... https://m.media-amazon.com/images/I/61j3SEUjMJ... https://www.amazon.co.uk/dp/B09B8YWXDF 4.7 15308 21.99 False 0 Hi-Fi Speakers
In [17]:
df_read.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 2222742 entries, 0 to 2222741
Data columns (total 10 columns):
 #   Column             Dtype  
---  ------             -----  
 0   asin               object 
 1   title              object 
 2   imgUrl             object 
 3   productURL         object 
 4   stars              float64
 5   reviews            int64  
 6   price              float64
 7   isBestSeller       bool   
 8   boughtInLastMonth  int64  
 9   categoryName       object 
dtypes: bool(1), float64(2), int64(2), object(5)
memory usage: 154.7+ MB
In [18]:
df = df_read.copy()
df.isnull().sum()
Out[18]:
asin                 0
title                0
imgUrl               0
productURL           0
stars                0
reviews              0
price                0
isBestSeller         0
boughtInLastMonth    0
categoryName         0
dtype: int64
In [19]:
df.columns
columns_to_drop = ['asin', 'imgUrl', 'productURL']
df = df.drop(columns=columns_to_drop)

isBestSeller' is the column I want . True = 1, False = 0¶

In [20]:
df['isBestSeller'] = df['isBestSeller'].astype(int)
In [21]:
df.head(3)
Out[21]:
title stars reviews price isBestSeller boughtInLastMonth categoryName
0 Echo Dot (5th generation, 2022 release) | Big ... 4.7 15308 21.99 0 0 Hi-Fi Speakers
1 Anker Soundcore mini, Super-Portable Bluetooth... 4.7 98099 23.99 1 0 Hi-Fi Speakers
2 Echo Dot (5th generation, 2022 release) | Big ... 4.7 15308 21.99 0 0 Hi-Fi Speakers

Selling Volume In Last Month¶

In [22]:
df['boughtInLastMonth'].unique()
Out[22]:
array([    0,   600,  1000,   200,   100,    50,  2000,   800,   900,
         700,   500,   400,   300,  4000,  3000,  5000,  9000,  7000,
        6000, 50000, 20000, 10000,  8000, 30000, 40000])

Number of Categories¶

In [23]:
df['categoryName'].nunique()
Out[23]:
296

How Many Bestsellers?¶

In [24]:
# Count the occurrences of True and False in the 'isBestSeller' column
is_bestseller_count = df['isBestSeller'].value_counts()
In [25]:
# Create a DataFrame for the pie chart
is_bestseller_data = pd.DataFrame({'isBestSeller': is_bestseller_count.index, 'Count': is_bestseller_count.values})
In [26]:
# Define custom labels
custom_labels = {0: "Not Bestseller", 1: "Bestseller"}
In [27]:
# Create an interactive pie chart with custom labels
fig = px.pie(is_bestseller_data, names='isBestSeller', values='Count', title='isBestSeller Count')
In [28]:
# Show the interactive pie chart
fig.show()

How Many boughtInLastMonth?¶

In [29]:
# Count the occurrences in the 'boughtInLastMonth' column
boughtInLastMonth_count = df['boughtInLastMonth'].value_counts()
In [30]:
# Combine values after the first 5 into 'Others' because they are too few to show.
top_5_counts = boughtInLastMonth_count[:5]
other_counts = boughtInLastMonth_count[5:]
In [31]:
# Sum the counts of the 'Others'
other_total = other_counts.sum()
In [32]:
# Create a DataFrame for the 'Others' category
others_data = pd.DataFrame({'boughtInLastMonth': ['Others'], 'Count': [other_total]})
In [33]:
# Create an interactive pie chart
fig = px.pie(boughtInLastMonth_data, names='boughtInLastMonth', values='Count', title='boughtInLastMonth Count')

---------------------------------------------------------------------------
NameError                                 Traceback (most recent call last)
Cell In[33], line 2
      1 # Create an interactive pie chart
----> 2 fig = px.pie(boughtInLastMonth_data, names='boughtInLastMonth', values='Count', title='boughtInLastMonth Count')

NameError: name 'boughtInLastMonth_data' is not defined
In [34]:
# Show the interactive pie chart
fig.show()
In [ ]: